Browse Items (14428 total)
Sort by:
-
Hybrid Machine Learning Approach for Gait Type Classification Using Pose-Based Feature Extraction
Gait analysis is essential for the diagnosis of neuromuscular and musculoskeletal disorders. Traditional methods are vulnerable and lead to inconsistency as they rely on subjective assessments. An angle-based approach which uses advanced machine learning techniques have been used address this. Extracted joint angle measurements have been extracted from the video data using computer vision methods. The characteristics used in this research were used to train a hybrid model of a Random Forest classifier and a Fuzzy C-Means clustering algorithm. Random Forest model was used as it is stable and capable of dealing with intricate nonlinear relationships and Fuzzy C-Means was used as it can manage ambiguity in the data as well as overlapping class distributions. The results showed that the Random Forest classifier has a classification accuracy of 94.62%, which is better than the other models in distinguishing between normal and abnormal gait patterns. Fuzzy C-Means also shows high accuracy is capable of clustering various forms of gait and extracting detailed features in gait dynamics. Results suggest that integrating joint angle analysis with machine learning methods provides a credible tool for gait analysis, which can aid clinicians in the early detection and treatment of gait related disorders. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026. -
Hybrid Machine Learning Models for Crowdfunding Success Prediction
Crowdfunding has become a viable option for founders and entrepreneurs as an alternative source of funding, where individuals can access a large pool of supporters to fill the funding gap. There are a variety of reasons why it is difficult to predict if crowdfunding project will be successful. As such, types of projects, duration of the campaign, target funding goal, and overall supporter activity are constantly changing. This research, therefore, aims to explore, the use of machine learning for predictive models that quantitatively leverage the historical records of projects on kickstarter, in order to find the success probability. In order to analyze the predictive ability for campaign success, was used a combination of machine learning models - Logistic Regression, Random Forest, XGBoost, LightGBM, and Decision Trees - the models that had the greatest precision were Decision Tree (99.91% acc), and LightGBM (99.90% acc) hence why they were selected. In addition, this research demonstrates how feature selection coupled with ensemble learning can significantly increase predictive potential by providing valuable information for campaign builders, platform operators, and investors who are undertaking crowdfunding projects. These findings indicate that predictive modelling can support campaign design, promote investor trust, and enhance credibility for crowdfunding platforms, through uncovering fraud. Additional measures measuring social media interaction or sentiment analysis could be incorporated to provide information for better predictive models. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026. -
Hybrid Mobile-Spinalnet with feature extraction for brain tumor detection using MRI images
Brain tumors are deadly and can hinder the normal functioning of the human body. Generally, surgical methods are preferred for treating brain tumors. Early and accurate detection remains a problem, due to the complexity of tumor shapes and poor generalization to diverse tumor types. To address this, a hybrid Mobile-SpinalNet is established in this paper for the detection of brain tumors with Magnetic Resonance Imaging (MRI) images. This system involves seven stages, including input image acquisition, image preprocessing, skill stripping, tumor segmentation, data augmentation, feature extraction, and brain tumor detection. Initially, image acquisition is carried out, and then the input image is preprocessed by using a Mean filter. Subsequently, the skull stripping is performed using Fuzzy C-Means (FCM). After that, by using the TransUNet, the tumor region is isolated in the segmentation module. Furthermore, the data augmentation is carried out, and then the feature mining takes place in the feature extraction phase to excerpt features such as Speeded Up Robust Features (SURF), Oriented Fast and Rotated BRIEF (ORB), Fuzzy Local Binary Pattern (FLBP) and statistical features. At last, the brain tumor is identified by employing a hybrid Mobile-SpinalNet. This framework fuses the MobileNet and SpinalNet depending on regression modeling with applied Fractional Calculus (FC). The Mobile-SpinalNet is validated for its efficacy by comparing it to other techniques, and it showed better performance with a precision of 0.953, accuracy of 0.943, and recall of 0.970. 2025 Elsevier Ltd -
Hybrid Model Integrating Firefly Algorithm and Gradient Boosted Trees for Stronger Financial Fraud Detection
Financial fraud poses a critical threat, with complex attack patterns overwhelming traditional rule-based detection methods. This project proposes a hybrid model integrating the Firefly Algorithm (FA) for hyperparameter optimization with Gradient Boosted Trees (GBTs) to enhance fraud detection. Evaluated on an imbalanced financial dataset, the FA-GBT framework achieved significant performance gains. ROC-AUC increased by 5.08% to 0.9883, demonstrating superior discrimination. Critically, Precision rose by 13.38%, substantially reducing false positives/false alarms and operational costs. The optimization also reduced training time by 64%(from 9.8 seconds to 3.5 seconds). This supports a robust, high-confidence, and efficient strategy for deployment in real-time transaction systems. 2025 IEEE. -
Hybrid Model Using Interacted-ARIMA andANN Models forEfficient Forecasting
When two models applied to the same dataset produce two different sets of forecasts, it is a good practice to combine the forecasts rather than using the better one and discarding the other. Alternatively, the models can also be combined to have a hybrid model to obtain better forecasts than the individual forecasts. In this paper, an efficient hybrid model with interacted ARIMA (INTARIMA) and ANN models is proposed for forecasting. Whenever interactions among the lagged variables exist, the INTARIMA model performs better than the traditional ARIMA model. This is validated through simulation studies. The proposed hybrid model combines forecasts obtained through the INTARIMA model from the dataset, and those through the ANN model from the residuals of INTARIMA, and produces better forecasts than the individual models. The quality of the forecasts is evaluated using three error metrics viz., Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Empirical results from the application of the proposed model on the real dataset - lynx - suggest that the proposed hybrid model gives superior forecasts than either of the individual models when applied separately. The methodology is replicable to any dataset having interactions among the lagged variables.. 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
Hybrid models for intraday stock price forecasting based on artificial neural networks and metaheuristic algorithms
Stock market prediction is one of the critical issues in fiscal market. It is important issue for the traders and investors. Artificial Neural Networks (ANNs) associated with nature inspired algorithms are playing an increasingly vital role in many areas including medical field, security systems and stock market. Several prediction models have been developed by researchers to forecast stock market trend. However, few studies have focused on improving stock market prediction accuracy especially when utilizing artificial neural networks to perform the analysis. This paper proposed nine new integrated models for forecasting intraday stock price based on the potential of three ANNs, Back Propagation Neural Network (BPNN), Radial Basis Function Neural Network (RBFNN), Time Delay Neural Network (TDNN) and nature inspired algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC).The developed models were named as GA-BPNN, PSO-BPNN, ABC-BPNN, GA-RBFNN, PSO-RBFNN, ABC-RBFNN, GA-TDNN, PSO-TDNN and ABC-TDNN. Nature inspired algorithms are employed for optimizing the parameters of ANNs. Technical indicators calculated from historical data are fed as input to developed models. Proposed hybrid models validated on four datasets representing different sectors in NSE. Four statistical metrics, Root Mean Square Error (RMSE), Hit Rate (HR), Error Rate (ER) and prediction accuracy were utilized to gauge the performance of the developed models. Results proved that the PSO-BPNN model yielded the highest prediction accuracy in estimating intraday stock price. The other models, GA-BPNN, ABC-BPNN, GA-RBFNN, PSO-RBFNN, ABC-RBFNN, GA-TDNN, PSO-TDNN and ABC-TDNN produced lower performance with mean prediction accuracy of 97.24%, 98.37%, 84.01%, 85.15%, 84.01%, 83.87%, 89.95% and 78.61% respectively. 2021 -
Hybrid nanofluid flow over a vertical rotating plate in the presence of hall current, nonlinear convection and heat absorption
An exact analysis has been carried out to study a problem of the nonlinear convective flow of hybrid nanoliquids over a vertical rotating plate with Hall current and heat absorption. Three different fluids namely CuAl2O3H2Ohybrid nanofluid, Al2O3H2O nanofluid and H2O basefluids are considered in the analysis. The simulation of the flow was carried out using the appropriate values of the empirical shape factor for five different particle shapes (i.e., sphere, hexahedron, tetrahedron, column and lamina). The governing PDEs with the corresponding boundary conditions are non-dimensionalised with the appropriate dimensionless variables and solved analytically by using LTM (Laplace transform technique). This investigation discusses the effects of governing parameters on velocity and temperature fields in addition to the rate of heat transfer. The numeric data of the density, thermal conductivity, dynamic viscosity, specific heat, Prandtl number and Nusselt number for twelve different hybrid nanofluids at 300 K is presented. The temperature profile of hybrid nanoliquid is larger than nanoliquid for same volume fraction of nanoparticles. Also, the glycerin-based nanoliquid has a high rate of heat transfer than engine oil, ethylene glycol and water-based nanoliquids in order. 2018 by American Scientific Publishers All rights reserved. -
Hybrid optimization for efficient 6G IoT traffic management and multi-routing strategy
Efficient traffic management solutions in 6G communication systems face challenges as the scale of the Internet of Things (IoT) grows. This paper aims to yield an all-inclusive framework ensuring reliable air pollution monitoring throughout smart cities, capitalizing on leading-edge techniques to encourage large coverage, high-accuracy data, and scalability. Dynamic sensors deployed to mobile ad-hoc pieces of fire networking sensors adapt to ambient changes. To address this issue, we proposed the Quantum-inspired Clustering Algorithm (QCA) and Quantum Entanglement and Mobility Metric (MoM) to enhance the efficiency and stability of clustering. Improved the sustainability and durability of the network by incorporating Dynamic CH selection employing Deep Reinforcement Learning (DRL). Data was successfully routed using a hybrid Quantum Genetic Algorithm and Ant Colony Optimization (QGA-ACO) approach. Simulation results were implemented using the ns-3 simulation tool, and the proposed model outperformed the traditional methods in deployment coverage (95%), cluster stability index (0.97), and CH selection efficiency (95%). This work is expected to study the 6G communication systems as a key enabler for IoT applications and as the title legible name explains, the solutions smartly done in a practical and scalable way gives a systematic approach towards solving the IoT traffic, and multi-routing challenges that are intended to be addressed in 6G era delivering a robust IoT ecosystem in securing the process. The Author(s) 2024. -
Hybrid Quantum Network with Snow Geese-Elk Herd Optimization for Smart Load Shedding in Grids with Electric Vehicles and Photovoltaic Systems
The increasing penetration of variable renewable energy and the growth of electric vehicles (EV) have created an urge for more sophisticated load management methods to ensure grid stability. Conventional load shedding (LS) methods are typically not equipped to manage the unpredictability brought about by these modern additions to the grid. This study introduces an innovative smart load-shedding strategy that uses a hybrid optimization model. At its core is a Quantum Neural Network (QNN), which enables intelligent and data-based load prioritization by evaluating factors such as load criticality, energy usage, responsiveness to demand, and operational flexibility. The required LS amount is calculated through a combined use of Snow Geese Optimization (SGO) and the Elk Herd Optimizer (EHO), with specific attention given to the flexibility offered by EVs to address the variability in photovoltaic (PV) power generation. Testing has been performed on the IEEE 33-bus network reveal a notable decrease in total load demand by around 33%, contributing to improved grid stability, with voltage levels staying close to 0.99 p.u. Additionally, the average load across the network buses dropped by roughly 52%. This hybrid approach not only ensures better performance but also achieves quicker convergence compared to existing optimization methods. The proposed intelligent LS method presents an effective strategy for preserving grid stability amid growing integration of renewables and EV by incorporating QNN with SGO and EHO while accounting for EV adaptability. The Author(s), under exclusive licence to Shiraz University 2025. -
Hybrid Renewable Road Side Charging Station with I2V Communication Functionality
The faster adoption of Renewable-based Energy Sources for charging Electric Vehicles is highly required. The paper proposes a novel strategy of design and developing a hybrid Road Side Unit (RSU) that would be easy to install and provides easy access to Electric Vehicle charging. The system inculcates Infrastructure to Vehicle (I2V) communication framework enabling communication between the Infrastructure and the Vehicle to identify the nearest charging station based on the availability. The communication framework is based on Wi-Fi communication and enables bidirectional communication between the Vehicle and the Infrastructure as well. The modelling and development of the RSU, and the active power flow regulation from the RSU to the Charging Station is also developed, using a Fuzzy Controller. 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Hybrid Renewable Source Powered Dual Input Single Output Converter With High Voltage Gain for Rural Healthcare Facilities
Rural dwellers need a well-equipped healthcare service for a decent life. Most of the rural areas located in southern parts of India away from the grid connection thereby lack in electricity. Unreliable electric power leads to the limited access or inaccessibility of most essential medical equipment in the clinic. The deficiency has also reduced rural healthcare centers ethics criteria. This research work finds all the available resources in the rural healthcare clinic and proposes hybrid solar PV source and supercapacitor-based approaches to make sure of reliable energy access and uninterrupted power supply. Any healthcare facilities include an emergency room, waiting hall, nursing room, consulting room, delivery room, male and female room, and a testing lab. It may take a daily average energy consumption of 16 kWh with 3 kW peak demand. In the input side, solar PV system with an H-type clamped capacitorbased boost converter is proposed for the reduction of input current ripples and power switch conduction losses. At the load side, a capacitor with a switch (switched capacitor) is considered to reduce voltage stress of the components present in the topology and to attain high gain. This research work adopts interleaved structure-based capacitors for current ripple reduction, and the series structure is considered to attain high gain. The proposed novel converter takes the input voltage of 40 V and produces the output voltage of 280 V. The DC link output is then connected with the voltage source inverter (VSI) to get a desired output. The proposed novel converter is employed to run a 3? induction motor for the AC load with the rating of 400 V, 15 A AC power. MATLAB R2015a software is preferred for the simulation analysis. Copyright 2025 K. M. D. Riyaz Ali et al. Journal of Electrical and Computer Engineering published by John Wiley & Sons Ltd. -
Hybrid response surface methodologyparticle swarm optimization framework for predictive modeling and tensile strength optimization of PLA bio-composites
Developing mechanically robust biodegradable composites is critical for next-generation orthopedic support devices. Although polylactic acid (PLA) is widely used in additive manufacturing, incorporating fillers can lead to reduced tensile performance when interfacial bonding with the matrix is inadequate. This study aims to enhance the mechanical performance of 3D-printed PLA reinforced with 2wt% rice husk-derived silica (SiO2) through optimized post-annealing. A hybrid statisticalcomputational framework combining Response Surface Methodology (RSM) and Particle Swarm Optimization (PSO) was implemented to identify optimal print speed, annealing temperature, and annealing time. PSO predicted the optimal conditions as 50mm/s, 90.90C, and 60min, respectively, corresponding to a projected ultimate tensile strength (UTS) of 54.65MPa. Confirmation experiments validated the prediction, yielding a mean UTS of 54.49MPa with an error below 1%. Scanning electron microscopy revealed improved interlayer fusion and enhanced ductility in the optimized samples relative to unoptimized ones. Overall, the integration of RSM and PSO effectively refined post-annealing conditions without modifying material composition, demonstrating a viable strategy for strengthening PLA-based biocomposites. The proposed framework provides a practical route for tailoring mechanical properties in biomedical additive manufacturing, particularly for load-bearing orthopedic applications. The Author(s) 2026 -
Hybrid scheme image compression using DWT and SVD
Image compression is process of reducing data size to represent an image by removing redundant data. Hybrid scheme image compression is combination of methods performed in order or as an amalgam to form a new technique. In this paper, we proposed a new approach to compress the image by collaborating Discrete Wavelet Transformation (DWT) and Singular Value Decomposition (SVD). Image is decomposed into wavelets using DWT and approximate wavelet is subsequently transformed into four bands. Different wavelet filters are implemented for transformation namely Haar, Daubechies, Biorthogonal and Coiflets. Apart from approximate image, SVD is applied on remaining wavelets (Horizontal, Vertical and Diagonal Details) at each decomposition level. On reconstruction, various singular values are selected depending on the level transformation. The performance of the proposed method is compared and evaluated with SVD, DCT-SVD and DWT-DCT-SVD. Evaluation is carried out based on Compression Ratio (CR), Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) index. From the experimental results, it is observed that proposed method yields better MSE, PSNR and SSIM compared to state-of-the-art methods. 2017, Institute of Advanced Scientific Research, Inc. All rights reserved. -
Hybrid Semantic Evaluation of Student Answers Using Rule Matching and BERT Embeddings
Accurate evaluation of student answers in online and traditional assessments is critical in education. In recent years, various text similarity-based methods have been proposed. However, there are certain challenges, such as the semantic and structural understanding of text. Thus, this paper uses the BERT model to present a hybrid evaluation framework that combines rule-based similarity techniques with deep semantic knowledge. The rule-based component utilizes predefined linguistic and domain-specific rules to ensure interpretability. At the same time, BERT-based similarity captures the semantic similarity and the paraphrased answers. Experimentation has been carried out on benchmark datasets, with the proposed hybrid model and human experts on manual evaluation. The performance comparison demonstrates that the hybrid model has performed significantly better than the traditional machine learning approaches in terms of accuracy and fairness of scoring. The proposed hybrid model is also compatible with deployment in educational platforms as it provides suitable feedback to learners. 2025 IEEE. -
Hybrid short term load forecasting using ARIMA-SVM
In order to perform a stable and reliable operation of the power system network, short term load forecasting is vital. High forecasting accuracy and speed are the two most important requirements of short-term load forecasting. It is important to analyze the load characteristics and to identify the main factors affecting the load. ARIMA method is most commonly used, as it predict the load purely based on the historical loads and no other assumptions are considered. Therefore there is a need for Outlier detection and correction method as the prediction is based on historical data, the historical data may contain some abnormal or missing values called outliers. Also the load demand is influenced by several other external factors such as temperature, day of the week etc., the Artificial Intelligence techniques will incorporate these external factors which improves the accuracy further. In this paper a hybrid model ARIMA-SVM is used to predict the hourly demand. ARIMA is used to predict the demand after correcting the outliers using Percentage Error (PE) method and its deviation is corrected using SVM. Main objective of this method is to reduce the Mean Absolute percentage Error (MAPE) by introducing a hybrid method employing with outlier detection technique. The historical load data of 2014-2015 from a utility system of southern region is taken for the study. It is observed that the MAPE error got reduced and its convergence speed increased. 2017 IEEE. -
Hybrid shuffled frog leaping and improved biogeography-based optimization algorithm for energy stability and network lifetime maximization in wireless sensor networks
Wireless sensor networks are significantly used for data sensing and aggregating dusts from a remote area environment in order to utilize them in a diversified number of engineering applications. The data transfer among the sensor nodes is attained through the inclusion of energy efficient routing protocols. These energy efficient routing necessitates optimal cluster head selection procedure for handling the challenge of energy consumption to extend the stability and lifetime in the sensor networks. The implementation of energy efficient routing is still complicated even when the process of clustering is enhanced through the cluster head selection. The majority of the existing cluster head selection schemes suffer from the issues of poor selection accuracy, increased computation, and duplicate nodes' selection. In this paper, hybrid shuffled frog leaping and improved biogeography-based optimization algorithm (HSFLBOA) for optimal cluster head selection is proposed for resolving issues that are common in cluster head selection schemes. This proposed HSFLBOA used the objective function that used the parameters of node energy, data packet transmission delay, cluster traffic density, and internode distance in the cluster. The simulation results of the proposed HSFLBOA is determined to be significant in achieving superior throughput and network energy compared to benchmarked metaheuristic optimal cluster head schemes. 2021 John Wiley & Sons Ltd. -
Hybrid sparse and block-based compressive sensing algorithm for industry based applications
Image reconstructions are a challenging task in MRI images. The performance of the MRI image can be measure by following parameters like mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Compromising the above parameters and reconstructing the MRI image leads to false diagnosing. To avoid the false diagnosis, we have combined sparse based compressive sensing and block-based compressive sensing algorithm, and we introduced the hybrid sparse and block-based compressive sensing algorithm (HSBCS). In compressive stage, however, image reconstruction performance is decreased, hence, in the image reconstruction module, we have introduced convex relaxation algorithm. This proposed algorithm is obtained by relaxing some of the constraints of the original problem and meanwhile extending the objective function to the larger space. The performance is compared with the existing algorithm, block-based compressive sensing algorithm (BCS), BCS based on discrete wavelet transform (DWT), and sparse based compress-sensing algorithm (SCS). The experimentation is carried out using BRATS dataset, and the performance of image compression HSBCS evaluated based on SSIM, and PSNR, which attained 56.19 dB, and 0.9812. Copyright 2024 Inderscience Enterprises Ltd. -
Hybrid Subset Feature Selection and Importance Framework
Feature selection algorithms are used in high-dimensional data to remove noise, reduce model overfitting, training and inference time, and get the importance of features. Features subset selection is choosing the subset with the best performance. This research provides a Hybrid subset feature selection and importance (HSFSI) framework that provides a pipeline with customization for choosing feature selection algorithms. The authors propose a hybrid algorithm in the HSFSI framework to select the best possible subset using an efficient exhaustive search. The framework is tested using the Bombay stock exchange IT index's companies' data collected quarterly for 16 years consisting of 71 financial ratios. The experimental results demonstrate that models created using 12 features chosen by the proposed algorithm outperform models with all features with up to 6% accuracy. The importance-based ranks of all features are generated using the framework calculated using 13 implemented feature selection techniques. All selected feature subsets are cross-validated using prediction models such as support vector machine, logistic regression, KNeighbors classffier, random forest, and deep neural network. The HSFSI framework is available as an open-source Python software package named ''feature-selectionpy'' available at GitHub and Python package index. 2023 IEEE. -
Hybridization of Texture Features for Identification of Bi-Lingual Scripts from Camera Images at Wordlevel
In this paper, hybrid texture features are proposed for identification of scripts of bi-lingual camera images for a combination of 10 Indian scripts with Roman scripts. Initially, the input gray-scale picture is changed over into an LBP image, then GLCM and HOG features are extracted from the LBP image named as LBGLCM and LBHOG. These two feature sets are combined to form a potential feature set and are submitted to KNN and SVM classifiers for identification of scripts from the bilingual camera images. In all 77,000-word images from 11 scripts each contributing 7000-word images. The experimental results have shown the identification accuracy as 71.83 and 71.62% for LBGLCM, 79.21 and 91.09% for LBHOG, and 84.48 and 95.59% for combined features called CF, respectively for KNN and SVM. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Hycons Renewable Private Limited: decision to accept or reject an equity investment
Learning outcomes: This study will help students determine the economic value of a firm particularly in case of a small business. The crux of the case is to help students estimate an enterprise value for a company and figure the actual worth of the company to aid in decision-making. Case overview/Synopsis: This case is about a decision dilemma faced by Shashi Hegde, Director, Hycons Renewable Private Ltd, a company ventured into the production of Bio-CNG. It is about a recent proposal received by the firm from APL Ltd for equity investment with 40% stake in the firm. The case reflects the dilemma faced by small businesses to choose between investment or loss of control. Accepting the proposal will bring in additional funds, whereas the Board loss control on the firm. The case revolves around this dilemma. To help Hegde in this task, he seeks advice from his CFO and his confidant Kumar. Complexity academic level: This case is most appropriate for a core finance class for both under-graduate and graduate programs. Supplementary materials: Teaching notes are available for educators only. Subject code: CSS 1: Accounting and Finance. 2022, Emerald Publishing Limited.